drm/amdkfd: Fix GPU mappings for APU after prefetch
Fix the following corner case:-
Consider a 2M huge page SVM allocation, followed by prefetch call for
the first 4K page. The whole range is initially mapped with single PTE.
After the prefetch, this range gets split to first page + rest of the
pages. Currently, the first page mapping is not updated on MI300A (APU)
since page hasn't migrated. However, after range split PTE mapping it not
valid.
Fix this by forcing page table update for the whole range when prefetch
is called. Calling prefetch on APU doesn't improve performance. If all
it deteriotes. However, functionality has to be supported.
v2: Use apu_prefer_gtt as this issue doesn't apply to APUs with carveout
VRAM
v3: Simplify by setting the flag for all ASICs as it doesn't affect dGPU
v4: Remove v2 and v3 changes. Force update_mapping when range is split
at a size that is not aligned to prange granularity
Suggested-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Philip Yang<Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:45 +0000 (16:57 +0100)]
drm/amdgpu/vce1: Workaround PLL timeout on FirePro W9000
Sometimes the VCE PLL times out waiting for CTLACK/CTLACK2.
When it happens, the VCE still works, but much slower.
Observed on a Tahiti GPU, but not all:
- FirePro W9000 has the issue
- Radeon R9 280X not affected
- Radeon HD 7990 not affected
As a workaround, on the affected chip just don't put the
VCE PLL in sleep mode. Leaving the VCE PLL in bypass mode
or reset mode both work. Using bypass mode is simpler.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:44 +0000 (16:57 +0100)]
drm/amdgpu/vce1: Enable VCE1 on Tahiti, Pitcairn, Cape Verde GPUs
Add the VCE1 IP block to the SI GPUs that have it.
Advertise the encoder capabilities corresponding to VCE1,
so the userspace applications can detect and use it.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Co-developed-by: Alexandre Demers <alexandre.f.demers@gmail.com> Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:43 +0000 (16:57 +0100)]
drm/amd/pm/si: Hook up VCE1 to SI DPM
On SI GPUs, the SMC needs to be aware of whether or not the VCE1
is used. The VCE1 is enabled/disabled through the DPM code.
Also print VCE clocks in amdgpu_pm_info.
Users can inspect the current power state using:
cat /sys/kernel/debug/dri/<card>/amdgpu_pm_info
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:42 +0000 (16:57 +0100)]
drm/amdgpu/vce1: Ensure VCPU BO is in lower 32-bit address space (v3)
Based on research and ideas by Alexandre and Christian.
VCE1 actually executes its code from the VCPU BO.
Due to various hardware limitations, the VCE1 requires
the VCPU BO to be in the low 32 bit address range.
However, VRAM is typically mapped at the high address range,
which means the VCPU can't access VRAM through the FB aperture.
To solve this, we write a few page table entries to
map the VCPU BO in the GART address range. And we make sure
that the GART is located at the low address range.
That way the VCE1 can access the VCPU BO.
v2:
- Adjust to v2 of the GART helper commit.
- Add empty line to multi-line comment.
v3:
- Instead of relying on gmc_v6 to set the GART space before GTT,
add a new function amdgpu_vce_required_gart_pages() which is
called from amdgpu_gtt_mgr_init() directly.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Co-developed-by: Alexandre Demers <alexandre.f.demers@gmail.com> Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:41 +0000 (16:57 +0100)]
drm/amdgpu/vce1: Implement VCE1 IP block (v2)
Implement the necessary functionality to support the VCE1.
This implementation is based on:
- VCE2 code from amdgpu
- VCE1 code from radeon (the old driver)
- Some trial and error
A subsequent commit will ensure correct mapping for
the VCPU BO, which will make this actually work.
v2:
- Use memset_io more.
- Use memcpy_toio more.
- Remove __func__ from warnings.
- Don't reserve and map the VCPU BO anymore.
- Add empty line to multi-line comments
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Co-developed-by: Alexandre Demers <alexandre.f.demers@gmail.com> Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:40 +0000 (16:57 +0100)]
drm/amdgpu/vce1: Load VCE1 firmware
Load VCE1 firmware using amdgpu_ucode_request, just like
it is done for other VCE versions.
All SI chips share the same VCE1 firmware file: vce_1_0_0.bin
which will be sent to linux-firmware soon.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Co-developed-by: Alexandre Demers <alexandre.f.demers@gmail.com> Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:39 +0000 (16:57 +0100)]
drm/amdgpu/vce1: Clean up register definitions
The sid.h header contained some VCE1 register definitions, but
they were using byte offsets (probably copied from the old radeon
driver). Move all of these to the proper VCE1 headers and ensure
they are in dword offsets.
Also add the register definitions that we need for the
firmware validation mechanism in VCE1.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Co-developed-by: Alexandre Demers <alexandre.f.demers@gmail.com> Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The VCPU BO doesn't only contain the VCE firmware but also other
ranges that the VCE uses for its stack and data. Let's initialize
this to zero to avoid having garbage in the VCPU BO.
Additionally, don't unmap/unreserve the VCPU BO.
The VCPU BO needs to stay at the same location before and after
sleep/resume because the FW code is not relocatable once it's
started.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:37 +0000 (16:57 +0100)]
drm/amdgpu/vce: Move firmware load to amdgpu_vce_early_init
Try to load the VCE firmware at early_init.
When the correct firmware is not found, return -ENOENT.
This way, the driver initialization will complete even
without VCE, and the GPU will be functional, albeit
without video encoding capabilities.
This is necessary because we are planning to add support
for the VCE1, and AMD hasn't yet publised the correct
firmware for this version. So we need to anticipate that
users will try to boot amdgpu on SI GPUs without the
correct VCE1 firmware present on their system.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:36 +0000 (16:57 +0100)]
drm/amdgpu/ttm: Use GART helper to map VRAM pages (v2)
Use the GART helper function introduced in the previous commit
to map the VRAM pages of the transfer window to GART.
No functional changes, just code cleanup.
Split this into a separate commit to make it easier to bisect,
in case there are problems in the future.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jonathan Kim [Thu, 6 Nov 2025 15:17:06 +0000 (10:17 -0500)]
drm/amdkfd: relax checks for over allocation of save area
Over allocation of save area is not fatal, only under allocation is.
ROCm has various components that independently claim authority over save
area size.
Unless KFD decides to claim single authority, relax size checks.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Philip Yang <philip.yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Sun, 9 Nov 2025 18:26:40 +0000 (19:26 +0100)]
drm/amdgpu: Use DC by default on SI dGPUs
Now that DC supports analog connectors, it has reached feature
parity with the legacy non-DC display driver on SI dGPUs.
Use the DC display driver by default on SI dGPUs, unless it is
explicitly disabled using the amdgpu.dc=0 module parameter.
DC brings proper support for DP/HDMI audio, DP MST,
10-bit colors, some HDR features, atomic modesetting, etc.
Also clarify the comment about what is missing to have full
DC support for CIK APUs.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:35 +0000 (16:57 +0100)]
drm/amdgpu/gart: Add helper to bind VRAM pages (v2)
Binds pages that located in VRAM to the GART page table.
Useful when a kernel BO is located in VRAM but
needs to be accessed from the GART address space,
for example to give a kernel BO a 32-bit address
when GART is placed in LOW address space.
v2:
- Refactor function to be more reusable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Fri, 7 Nov 2025 15:57:34 +0000 (16:57 +0100)]
drm/amdgpu/gmc6: Place gart at low address range
Instead of using a best-fit algorithm to determine which part
of the VMID 0 address space to use for GART, always use the low
address range.
A subsequent commit will use this to map the VCPU BO in GART
for the VCE1 IP block.
Split this into a separate patch to make it easier to bisect,
in case there are any errors in the future.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jesse.Zhang [Fri, 7 Nov 2025 11:19:08 +0000 (19:19 +0800)]
drm/amdgpu: resume MES scheduling after user queue hang detection and recovery
This patch ensures the Micro-Engine Scheduler (MES) is properly resumed
after detecting and recovering from a user queue hang condition.
Key changes:
1. Track when a hung user queue is detected using found_hung_queue flag
2. Call amdgpu_mes_resume() to restart MES scheduling after completing
the hang recovery process
3. This complements the existing recovery steps (fence force completion
and device wedging) by ensuring the scheduler can process new work
Without this resume call, the MES scheduler may remain in a paused state
even after the hung queue has been handled, preventing newly submitted
work from being processed and leading to system stalls.
Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Sathishkumar S [Tue, 7 Oct 2025 07:47:51 +0000 (13:17 +0530)]
drm/amdgpu/jpeg: Add parse_cs for JPEG5_0_1
enable parse_cs callback for JPEG5_0_1.
Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Asad Kamal [Thu, 6 Nov 2025 16:56:32 +0000 (00:56 +0800)]
drm/amd/pm: Remove power2_average node
SOC power consumption is reported by power1_average.
power2_cap_default/min/max only represent second level limits
and don't represent a different type of power or power consumption
by a subsection of the SOC. Therefore power2_average does not serve any
purpose and hence removing power2_average sysfs node
Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/display: Add kdoc params/returns in dc/link detection helpers
The link detection helpers in dc/link/link_detection.c were missing
kdoc annotations for parameters and return values.
Fixes the below with gcc W=1:
...link_detection.c:872 parameter 'edid_header' not described
...link_detection.c:890 parameter 'link' not described
...link_detection.c:914 parameter 'link' not described
...link_detection.c:1355 parameter 'link' not described
...link_detection.c:1355 parameter 'type' not described
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Cc: Roman Li <roman.li@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/display: Fix annotations for connector poll/detect parameters
Adds the missing @aconnector, @connector, and @force descriptions:
@aconnector – This is the DM (Display Manager) connector. It gives
access to the DRM connector, the DC link, and hotplug/poll state. The
code uses it to check the link, update the sink, and manage connector
state changes.
@connector – This is the main DRM connector given by the DRM core.
Inside the detect function, it is converted to amdgpu_dm_connector so we
can run DC link detection, either light or full.
@force – This flag tells the function whether to run a full detect
again. If false, we avoid heavy DAC load detect steps to prevent
flicker. If true, we force a re-detect even when we normally skip it.
Fixes the below with gcc W=1:
function param 'aconnector' not described in 'amdgpu_dm_connector_poll'
function param 'force' not described in 'amdgpu_dm_connector_poll'
function param 'connector' not described in 'amdgpu_dm_connector_detect'
function param 'force' not described in 'amdgpu_dm_connector_detect'
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Cc: Roman Li <roman.li@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Sultan Alsawaf [Fri, 7 Nov 2025 18:07:13 +0000 (13:07 -0500)]
drm/amd/amdgpu: Ensure isp_kernel_buffer_alloc() creates a new BO
When the BO pointer provided to amdgpu_bo_create_kernel() points to
non-NULL, amdgpu_bo_create_kernel() takes it as a hint to pin that address
rather than allocate a new BO.
This functionality is never desired for allocating ISP buffers. A new BO
should always be created when isp_kernel_buffer_alloc() is called, per the
description for isp_kernel_buffer_alloc().
Ensure this by zeroing *bo right before the amdgpu_bo_create_kernel() call.
Fixes: 55d42f616976 ("drm/amd/amdgpu: Add helper functions for isp buffers") Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Pratap Nirujogi <pratap.nirujogi@amd.com> Signed-off-by: Sultan Alsawaf <sultan@kerneltoast.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Taimur Hassan [Sat, 1 Nov 2025 01:07:35 +0000 (20:07 -0500)]
drm/amd/display: Promote DC to 3.2.358
Summary:
* Enable VRR when unsynced with the stream
* Refactor DSC cap calculation for dcn35
* Add debug log for power feature
* Fix fill latency issue
* Do not initialize LSDMA if it is not supported by DMU
Taimur Hassan [Fri, 31 Oct 2025 23:00:47 +0000 (19:00 -0400)]
drm/amd/display: [FW Promotion] Release 0.1.35.0
Summary for changes in firmware:
* Use panel_inst instead of otg_inst when getting fw state
* Contrast strength improves when HDR desktop mode
* Ensure pipes have no outstanding HUBP requests prior to IPS RCG entry
* Check for vm request and vm idle status in IPS1/2 entry sequence
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Leo Chen [Fri, 31 Oct 2025 19:25:47 +0000 (15:25 -0400)]
drm/amd/display: dynamically clock gate before and after prefetch
[Why]
An invalidation request arriving during prefetch can potentially hang
the system if dynamic clock gating is enabled and memory power requests
are disabled.
[How]
• Disable clock gating and enable memory power requests for the duration
of the prefetch.
• Turn on clock gating and disable memory power requests again after
prefetch is complete.
Limit the scope for DCN35 and DCN42 only.
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Leo Chen <leo.chen@amd.com> Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Mohit Bawa [Thu, 23 Oct 2025 14:40:41 +0000 (10:40 -0400)]
drm/amd/display: refactor DSC cap calculation for dcn35
why:
dcn35 currently uses a hardcoded DSC display clock value which is incorrect
for some asic types. Newer DCN versions retrieve dsc display clock from
clk_mgr. The same can be done for dcn35.
how:
Refactor the DSC cap calculation using pre-existing logic.
Handle ODM combine requirements in dc_dsc.c.
Replace hardcoded display clock with actual value retrieved from clk_mgr.
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by: Charlene Liu <charlene.liu@amd.com> Reviewed-by: Wenjing Liu <wenjing.liu@amd.com> Signed-off-by: Mohit Bawa <Mohit.Bawa@amd.com> Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
George Shen [Mon, 6 Oct 2025 15:23:31 +0000 (11:23 -0400)]
drm/amd/display: Add interface to capture power feature status for debug logging
[Why]
The status of various power features is often important information when
debugging certain issues, such as underflow. This info helps to
narrow down the potential sources of errors.
[How]
Add dc interface to capture power feature enablement status.
Reviewed-by: Dillon Varone <dillon.varone@amd.com> Signed-off-by: George Shen <george.shen@amd.com> Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Ivan Lipski [Thu, 23 Oct 2025 14:03:59 +0000 (10:03 -0400)]
drm/amd/display: Allow VRR params change if unsynced with the stream
[Why]
When changing resolution (e.g., 4K → FHD) in mirror/clone mode with
certain monitors, the monitor blanks and loses connection due to an early
exit in vrr_settings_require_update(). The function only checks if VRR
state, fixed refresh target, or min/max refresh rate range has changed.
During mode changes, if the calculated min/max refresh values remain the
same even though the stream's v_total changed, the function returns early
without updating vrr_params.adjust.v_total_min/max, leaving the monitor's
VRR timing parameters unsynced with the new mode, causing it to blank out.
[How]
Explicitly adjust VRR parameters to the stream's nominal v_total when VRR
is supported, but inactive.
Fixes: 6d31602a9f57 ("drm/amd/display: more liberal vmin/vmax update for freesync") Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Ivan Lipski <ivan.lipski@amd.com> Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Gangliang Xie [Thu, 6 Nov 2025 03:06:21 +0000 (11:06 +0800)]
drm/amd/pm: remove unnecessary prints for smu busy
smu busy is a normal case when calling SMU_MSG_GetBadPageCount, so no need
to print error status at each time.Instead, only print error status when
timeout given by user is reached.
Tao Zhou [Thu, 6 Nov 2025 08:26:56 +0000 (16:26 +0800)]
drm/amdgpu: optimize timeout implemention in ras_eeprom_update_record_num
The busy status returned by ras_eeprom_update_record_num may not be
an error, increase timeout to exclude false busy status. Also add more
comments to make the code readable.
v2: define a macro for the timeout value.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jesse.Zhang [Fri, 24 Oct 2025 08:09:25 +0000 (16:09 +0800)]
drm/amdgpu: fix lock warning in amdgpu_userq_fence_driver_process
Fix a potential deadlock caused by inconsistent spinlock usage
between interrupt and process contexts in the userq fence driver.
The issue occurs when amdgpu_userq_fence_driver_process() is called
from both:
- Interrupt context: gfx_v11_0_eop_irq() -> amdgpu_userq_fence_driver_process()
- Process context: amdgpu_eviction_fence_suspend_worker() ->
amdgpu_userq_fence_driver_force_completion() -> amdgpu_userq_fence_driver_process()
In interrupt context, the spinlock was acquired without disabling
interrupts, leaving it in {IN-HARDIRQ-W} state. When the same lock
is acquired in process context, the kernel detects inconsistent
locking since the process context acquisition would enable interrupts
while holding a lock previously acquired in interrupt context.
Fix by using spin_lock_irqsave()/spin_unlock_irqrestore() to properly
manage interrupt state regardless of calling context.
Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm_sched_entity_init wasn't called yet, so the only thing to
do is to release allocated memory.
This doesn't fix any bug since entity is zero allocated and
drm_sched_entity_fini does nothing in this case.
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd: Clarify that amdgpu.audio only works for non-DC
The comment already explains it but the module parameter help text
doesn't.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4684 Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Vitaly Prosyak [Thu, 6 Nov 2025 17:35:53 +0000 (12:35 -0500)]
drm/amdgpu: disable peer-to-peer access for DCC-enabled GC12 VRAM surfaces
Certain multi-GPU configurations (especially GFX12) may hit
data corruption when a DCC-compressed VRAM surface is shared across GPUs
using peer-to-peer (P2P) DMA transfers.
Such surfaces rely on device-local metadata and cannot be safely accessed
through a remote GPU’s page tables. Attempting to import a DCC-enabled
surface through P2P leads to incorrect rendering or GPU faults.
This change disables P2P for DCC-enabled VRAM buffers that are contiguous
and allocated on GFX12+ hardware. In these cases, the importer falls back
to the standard system-memory path, avoiding invalid access to compressed
surfaces.
Future work could consider optional migration (VRAM→System→VRAM) if a
performance regression is observed when `attach->peer2peer = false`.
Tested on:
- Dual RX 9700 XT (Navi4x) setup
- GNOME and Wayland compositor scenarios
- Confirmed no corruption after disabling P2P under these conditions
v2: Remove check TTM_PL_VRAM & TTM_PL_FLAG_CONTIGUOUS.
v3: simplify for upsteam and fix ip version check (Alex)
Suggested-by: Christian König <christian.koenig@amd.com> Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Wenjing Liu [Thu, 6 Nov 2025 05:49:18 +0000 (13:49 +0800)]
drm/amd/display: add macros to simplify code
[Why & How]
Adding macros to simplify the process of adding new error codes.
Currently, to add an error code, the developer needs to add both the
enum and the string translation. This is error prone and can lead to
inconsistencies. The refactor adds a macro to automatically add the
string translation based on the enum.
Reviewed-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Wenjing Liu <wenjing.liu@amd.com> Signed-off-by: Ray Wu <ray.wu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Dave Airlie [Fri, 7 Nov 2025 02:40:51 +0000 (12:40 +1000)]
Merge tag 'drm-misc-next-2025-11-05-1' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next
drm-misc-next for v6.19-rc1:
UAPI Changes:
- Add userptr support to ivpu.
- Add IOCTL's for resource and telemetry data in amdxdna.
Core Changes:
- Improve some atomic state checking handling.
- drm/client updates.
- Use forward declarations instead of including drm_print.h
- RUse allocation flags in ttm_pool/device_init and allow specifying max
useful pool size and propagate ENOSPC.
- Updates and fixes to scheduler and bridge code.
- Add support for quirking DisplayID checksum errors.
Driver Changes:
- Assorted cleanups and fixes in rcar-du, accel/ivpu, panel/nv3052cf,
sti, imxm, accel/qaic, accel/amdxdna, imagination, tidss, sti,
panthor, vkms.
- Add Samsung S6E3FC2X01 DDIC/AMS641RW, Synaptics TDDI series DSI,
TL121BVMS07-00 (IL79900A) panels.
- Add mali MediaTek MT8196 SoC gpu support.
- Add etnaviv GC8000 Nano Ultra VIP r6205 support.
- Document powervr ge7800 support in the devicetree.
Dave Airlie [Thu, 6 Nov 2025 23:45:38 +0000 (09:45 +1000)]
Merge tag 'drm-intel-next-2025-11-04' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-next
drm/i915 feature pull for v6.19:
Features and functionality:
- Enable LNL+ content adaptive sharpness filter (CASF) (Nemesa)
- Use optimized VRR guardband (Ankit, Ville)
- Enable Xe3p LT PHY (Suraj)
- Enable FBC support for Xe3p_LPD display (Sai Teja, Vinod)
- Specify DMC firmware for display version 30.02 (Dnyaneshwar)
- Report reason for disabling PSR to debugfs (Michał)
- Extend i915_display_info with Type-C port details (Khaled)
- Log DSI send packet sequence errors and contents
Refactoring and cleanups:
- Refactoring to prepare for VRR guardband optimization (Ankit)
- Abstract VRR live status wait (Ankit)
- Refactor VRR and DSB timing to handle Set Context Latency explicitly (Ankit)
- Helpers for prefill latency calculations (Ville)
- Refactor SKL+ watermark latency setup (Ville)
- VRR refactoring and cleanups (Ville)
- SKL+ universal plane cleanups (Ville)
- Decouple CDCLK from state->modeset refactor (Ville)
- Refactor VLV/CHV clock functions (Jani)
- Refactor fbdev handling (Jani)
- Call i915 and xe runtime PM from display via function pointers (Jouni)
- IRQ code refactoring (Jani)
- Drop display dependency on i915 feature check macros (Jani)
- Refactor and unify i915 and xe stolen memory interfaces towards display (Jani)
- Switch to driver agnostic drm to display pointer chase (Jani)
- Use display version over graphics version in display code (Matt A)
- GVT cleanups (Jonathan, Andi)
- Rename a VLV clock function to unify (Michał)
- Explicitly sanitize DMC package header num entries (Luca)
- Remove redundant port clock check from ALPM (Jouni)
- Use sysfs_emit() instead of sprintf() in PMU sysfs (Madhur Kumar)
- Clean up C20 PHY PLL register macros (Imre, Mika))
- Abstract "address in MMIO table" helper for general use (Matt A)
- Improve VRR platform abstractions (Ville)
- Move towards more standard PCI PM code usage (Ville)
- Framebuffer refactoring (Ville)
- Drop display dependency on i915_utils.h (Jani)
- Include cleanups (Jani)
Fixes:
- Workaround docking station DSC issues with high pixel clock and bpp (Imre)
- Fix Panel Replay in DSC mode (Imre)
- Disable tracepoints for PREEMPT_RT as a workaround (Maarten)
- Fix intel_crtc_get_vblank_counter() on PREEMPT_RT (Maarten)
- Fix C10 PHY identification on PTL/WCL (Dnyaneshwar)
- Take AS SDP into account with optimized guardband (Jouni)
- Fix panic structure allocation memory leak (Jani)
- Adjust an FBC workaround platforms (Vinod)
- Add fallback for CDCLK selection (Naladala)
- Avoid using invalid transcoder in MST transport select (Suraj)
- Don't use cursor size reduction on display version 14+ (Nemesa)
- Fix C20 PHY PLL register programming (Imre, Mika)
- Fix PSR frontbuffer flush handling (Jouni)
- Store ALPM parameters in crtc state (Jouni)
- Defeature DRRS on LNL+ (Ville)
- Fix the scope of the large DRAM DIMM workaround (Ville)
- Fix PICA vs. AUX power ordering issue (Gustavo)
- Fix pixel rate for computing watermark line time (Ville)
- Fix framebuffer set_tiling vs. addfb race (Ville)
- DMC event handler fixes (Ville)
DRM Core:
- CRTC sharpness strength property (Nemesa)
- DPCD DSC quirk for Synaptics Panamera devices (Imre)
- Helpers to query the branch DSC max throughput/line-width (Imre)
Merges:
- Backmerge drm-next for v6.18-rc and to sync with drm-xe-next (Jani)
Asad Kamal [Thu, 6 Nov 2025 06:11:45 +0000 (14:11 +0800)]
drm/amd/pm: Update default power1_cap
Update default power1_cap to max limit for smu_v13_0_6 and smu_v13_0_12
Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Wayne Lin [Wed, 5 Nov 2025 02:36:31 +0000 (10:36 +0800)]
drm/amd/display: Enable mst when it's detected but yet to be initialized
[Why]
drm_dp_mst_topology_queue_probe() is used under the assumption that
mst is already initialized. If we connect system with SST first
then switch to the mst branch during suspend, we will fail probing
topology by calling the wrong API since the mst manager is yet to
be initialized.
[How]
At dm_resume(), once it's detected as mst branc connected, check if
the mst is initialized already. If not, call
dm_helpers_dp_mst_start_top_mgr() instead to initialize mst
V2: Adjust the commit msg a bit
Fixes: bc068194f548 ("drm/amd/display: Don't write DP_MSTM_CTRL after LT") Cc: Fangzhi Zuo <jerry.zuo@amd.com> Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Wayne Lin <Wayne.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tao Zhou [Wed, 23 Jul 2025 11:04:17 +0000 (19:04 +0800)]
drm/amdgpu: add ras_eeprom_read_idx interface
PMFW will manage RAS eeprom data by itself, add new interface to read
eeprom data via PMFW, we can read part of records by setting index.
v2: use IPID parse interface.
pa is not used and set it to a fixed value.
v3: optimize the null pointer check for IPID parse interface.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd: Fix suspend failure with secure display TA
commit c760bcda83571 ("drm/amd: Check whether secure display TA loaded
successfully") attempted to fix extra messages, but failed to port the
cleanup that was in commit 5c6d52ff4b61e ("drm/amd: Don't try to enable
secure display TA multiple times") to prevent multiple tries.
Add that to the failure handling path even on a quick failure.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4679 Fixes: c760bcda8357 ("drm/amd: Check whether secure display TA loaded successfully") Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Samuel Zhang [Wed, 5 Nov 2025 03:04:08 +0000 (03:04 +0000)]
drm/amdgpu: fix gpu page fault after hibernation on PF passthrough
On PF passthrough environment, after hibernate and then resume, coralgemm
will cause gpu page fault.
Mode1 reset happens during hibernate, but partition mode is not restored
on resume, register mmCP_HYP_XCP_CTL and mmCP_PSP_XCP_CTL is not right
after resume. When CP access the MQD BO, wrong stride size is used,
this will cause out of bound access on the MQD BO, resulting page fault.
The fix is to ensure gfx_v9_4_3_switch_compute_partition() is called
when resume from a hibernation.
KFD resume is called separately during a reset recovery or resume from
suspend sequence. Hence it's not required to be called as part of
partition switch.
Signed-off-by: Samuel Zhang <guoqing.zhang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
unref bo after amdgpu_bo_reserve() failure as it has
called amdgpu_bo_ref() already
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alok Tiwari [Fri, 17 Oct 2025 19:16:21 +0000 (12:16 -0700)]
drm: rcar-du: fix incorrect return in rcar_du_crtc_cleanup()
The rcar_du_crtc_cleanup() function has a void return type, but
incorrectly uses a return statement with a call to drm_crtc_cleanup(),
which also returns void.
Remove the return statement to ensure proper function semantics.
No functional change intended.
Karol Wachowski [Tue, 4 Nov 2025 13:24:18 +0000 (14:24 +0100)]
accel/ivpu: Improve debug and warning messages
Add IOCTL debug bit for logging user provided parameter validation
errors.
Refactor several warning and error messages to better reflect fault
reason. User generated faults should not flood kernel messages with
warnings or errors, so change those to ivpu_dbg(). Add additional debug
logs for parameter validation in IOCTLs.
Check size provided by in metric streamer start and return -EINVAL
together with a debug message print.
Lizhi Hou [Tue, 4 Nov 2025 06:25:43 +0000 (22:25 -0800)]
accel/amdxdna: Add hardware specific attributes
Add three hardware specific attributes to describe device capabilities:
hwctx_limit: The maximum number of hardware context supported.
max_tops: The maximum TOPS supported.
curr_tops: The TOPS achievable with the current power and frequency
configuration.
Alex Deucher [Wed, 22 Oct 2025 21:11:38 +0000 (17:11 -0400)]
drm/amdgpu: fix possible fence leaks from job structure
If we don't end up initializing the fences, free them when
we free the job. We can't set the hw_fence to NULL after
emitting it because we need it in the cleanup path for the
submit direct case.
v2: take a reference to the fences if we emit them
v3: handle non-job fence in error paths
Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com> (v1) Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
YiPeng Chai [Tue, 28 Oct 2025 08:18:31 +0000 (16:18 +0800)]
drm/amdgpu: suspend ras module before gpu reset
During gpu reset, all GPU-related resources are
inaccessible. To avoid affecting ras functionality,
suspend ras module before gpu reset and resume
it after gpu reset is complete.
V2:
Rename functions to avoid misunderstanding.
V3:
Move flush_delayed_work to amdgpu_ras_process_pause,
Move schedule_delayed_work to amdgpu_ras_process_unpause.
V4:
Rename functions.
V5:
Move the function to amdgpu_ras.c.
Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Gangliang Xie <ganglxie@amd.com> Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/pm: implement ras_smu_drv interface for smu v13.0.12
implement ras_smu_drv interface for smu v13.0.12
Signed-off-by: Gangliang Xie <ganglxie@amd.com> Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Philip Yang [Wed, 29 Oct 2025 13:41:04 +0000 (09:41 -0400)]
Revert "drm/amdkfd: Improve signal event slow path"
To fix regression report on gfx8, which requires the exhaustive search
path for signaled event.
The high CPU usage of KFD interrupt wq issue is gone after HIP/ROCr add
option to reduce HW event interrupts, safe to revert this optimization
patch now.
Signed-off-by: Philip Yang <Philip.Yang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rong Zhang [Mon, 13 Oct 2025 16:47:35 +0000 (00:47 +0800)]
drm/amd/display: Fix NULL deref in debugfs odm_combine_segments
When a connector is connected but inactive (e.g., disabled by desktop
environments), pipe_ctx->stream_res.tg will be destroyed. Then, reading
odm_combine_segments causes kernel NULL pointer dereference.
Philip Yang [Fri, 31 Oct 2025 14:50:02 +0000 (10:50 -0400)]
drm/amdkfd: Don't clear PT after process killed
If process is killed. the vm entity is stopped, submit pt update job
will trigger the error message "*ERROR* Trying to push to a killed
entity", job will not execute.
Suggested-by: Christian König <christian.koenig@amd.com> Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Mon, 18 Aug 2025 06:33:41 +0000 (12:03 +0530)]
drm/amd/pm: Add schema v1.1 for parition metrics
Use a schema similar to gpu metrics v1.9 for partition metrics also. It
will have field type encoded followed by the field value(s). The
attribute ids used will be shared with gpu metrics. The structure
definition is only to distinguish between gpu metrics and partition
metrics though both gpu metrics v1.9 and partition metrics v1.1 follow
the same definition.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
projects / thirdparty / kernel / linux.git / log
